After recording losses in Q2, bonds rebounded somewhat to show a positive performance in Q3. The All-Bond Index (ALBI) returned 2.1% for the quarter. Longer-dated bonds performed better than shorter-dated ones. Again, inflation-linked bonds (which are not included in the ALBI) were the best performing fixed interest asset class, returning 3.7% for the quarter.
Bond market performance was perhaps surprising for many market-watchers, given that we entered the quarter just following the first interest rate increase in this cycle in June, and ended it amid a volatile and much weaker rand. Even as expectations cemented around further rate rises - and despite a poor performance by the rand - bond yields generally moved in a sideways range over the quarter. However, shorter dated bonds were affected by expectations about further repo rate rises, and by the end of the quarter the yield spread between the R157 and the R194 had moved to a small inversion as the latter, shorter-dated bond yield rose above that of the longerdated one.
There are a number of possible explanations as to why longerdated bond yields did not respond more negatively, especially later in the quarter when the rand broke important levels again and at one point seemed close to breaking $/R8.00. We believe one of the reasons was that the sharp reaction in the bond market to, in quick succession, the announcements of the June rate hike and the large Q1 current account deficit effectively "priced the cycle" within a few weeks. As can be seen from the chart below, there had been a very close relationship between moves in the rand and bond yields for the preceding year. In late June, bond yields broke higher and stayed at higher levels - pointedly ignoring the later rally seen in the currency in July and August. It thus seems feasible that the reason that longer-dated bonds largely ignored the weakening in the rand during September was that they had effectively already discounted it.
Going beyond simply looking at what bonds might have discounted, we can also discuss the reasons why. The most likely explanation seems to be that the credibility of the SARB had been established through the 50 basis points rate hike in June, and the follow-up of similar magnitude in September. In other words, the bond market has confidence that the SARB will respond appropriately to ensure that any inflationary impact of the rand's weakness (and there will be an effect) will be shortlived. We can perhaps look to 2001/2002 as a precedent in this instance. Following the sharp weakening of the rand through R10 to the dollar in 2001 Q4 and its subsequent move to close to R14, bonds sold off in dramatic fashion; at that stage there were many in the market who did not believe that the SARB would raise rates. However, when the SARB did indeed start raising rates (beginning with an "emergency" hike at a special, non-scheduled MPC meeting in January 2002) the bond market relaxed somewhat and bond yields generally moved down over the course of 2002 despite rising repo rates and rising inflation. In other words, the bond market looked through the spike in inflation as being temporary, and we expect that this type of thinking is probably supporting the market at present as well.
However, it is also important to note that the downward move in bond yields over 2002 was generally accompanied by a strengthening rand as well; and the rand probably remains the key risk at present. Hovering in the high R7's in the midst of very volatile moves at the time of writing, the shorter-term path seems somewhat unclear at present. But with the sharp adjustment in the rand that has been seen this year (it is currently down around 25% on a trade-weighted basis so far this year) combining with the interest rate rises the SARB has delivered already and more to come, it is just a matter of time before the principal imbalance in the economy - the current account deficit - starts to correct. It is this that has been at the root of concerns about the rand. Thus, while the shorter term path may be murky and the risk always remains of further negative sentiment affecting the rand, we are pretty confident that looking 12-18 months out will put us in a situation where the current account deficit has started adjusting towards more manageable levels, and inflation will have pushed past its peak and begun returning to trend levels.
While much of our discussion this time has centred on domestic developments, the international backdrop remains of critical importance. The rand's reaction to the news of the current account deficit should be seen in the context of jitters in emerging market currencies - especially of countries with high current account deficits - around the same time. Furthermore, the restraint in local bond yields, while probably largely due to the reasons we outlined above, would probably not have been possible in the absence of two important global factors: namely, that US Treasury yields and emerging market bond spreads (as measured by the EMBI) both remained at relatively low levels during the quarter.

Mark le Roux
Portfolio Manager

Bonds extended their losses into the second quarter, with the All-Bond Index (ALBI) down 3.6% for the quarter. This brings the year-to-date loss on the ALBI to just over 2%. A defensive, short-duration position in the Bond Fund proved the right move as the fund handsomely outperformed the ALBI as the market sold off. Year-to-date the fund produced a negative 0.77% return versus the negative 2.1% of the ALBI. The longest maturity area of the ALBI was down some 6.4% in the quarter and 7.0% for the year so far.
Bonds are the worst-performing asset class for the year to date, with the negative performance being trounced by equities - the All-Share Index was up 4.9% for the quarter, despite the jitters seen in May and June, and still up a smart 18.8% for the year- to-date. Bonds also trailed cash (3.6% year to date) as well as inflation-linked bonds, 3.3% for the year so far.
Poor bond index performance in the second quarter was led by a number of factors, principally the reduction in risk appetite for emerging market assets that we had expected would eventuate sometime this year. The trigger point seemed to be the US Federal Reserve raising interest rates beyond where many had expected, coupled with expectations of further monetary tightening by both the European Central Bank and the Bank of Japan. In addition to these major central banks, many other central banks have been in, or have begun a, tightening mode as well, including a majority of emerging markets. Rising interest rates both reduce general liquidity available in the markets - usually meaning a fall in risk appetite - and raise the relative return required by other assets if they are to continue to remain attractive relative to "home" interest-bearing assets.
Although the JPMorgan Emerging Markets Bond Index (EMBI) has held up quite well, it is significantly weaker than its best levels of earlier in the year. Local bonds were affected by this as well as by the fall in the rand that accompanied the risk reduction. The trade-weighted rand has lost some 18% since late April. The cherry on top for the bond market was the SA Reserve Bank's move to raise the repo rate by 50 basis points to 7.5% at the June Monetary Policy Committee meeting. While the FRA market had discounted some probability of a rate rise, it seemed that the bond market was to a large extent caught unawares by the move. The initial response by many economists was to dismiss the rate move as a one-off, but a negative current account number for the first quarter saw a widespread revision of rate forecasts. We have for some time expected that interest rates would enter an upcycle this year and thus fortunately had been positioned for the market moves.
Of course, the big question is what happens next. Readers will recall that we have been relatively bearish on inflation for some time, as we had been negative on currency and food price prospects, in particular. While we thus are not implementing any material adjustments to our inflation forecast in light of recent developments, some factors have moved worse than we had expected and thus upside risks remain. At present we expect the SARB to raise interest rates another 100 basis points by February 2007.
While inflation-linked bonds have outperformed conventional bonds over the past year, we believe that an overweight in this asset class is no longer justified. Real yields had fallen to levels that do not look sustainable, while breakeven inflation rates have risen to levels that no longer offer value. Looking forward, it seems it will be very difficult for inflation linkers to outperform cash.
Another interesting development in the domestic market was a change in sentiment towards credit, where the previous "nobrainer" attitude towards new corporate issues received a shock when a couple of new issues struggled to find a decent reception in the market. We have for some time believed that credit spreads in general were too tight (i.e. they did not offer value) and that it made sense to be defensive in credit (i.e. both underweight credit and within that, to remain in high-quality issuer names). Any risk aversion that enters markets is likely to make its way down the credit curve, and this is what we saw in the past quarter.
While we had been defensively positioned in the run-up to the repo rate hike, we feel that the bond market is now closer to fair value. Still, the call between bonds and cash remains close over the next 12 months - especially on a risk-adjusted basis.

Mark le Roux
Portfolio Manager

In a relatively rare turn of events in recent times, the SA bond market produced a negative return in March. The All Bond Index (ALBI) fell by 0.23% over the month, although shorterdated bonds did produce small positive returns which, from an index perspective, were outweighed by losses in longer-dated bonds. Over the first quarter, bonds still produced a positive return of 1.5%. Again, the shorter area was the place to be: the 1 - 3 and 3 - 7 year areas each returned 1.8%, but longer bonds produced less and the 12-years plus area returned just 0.6% over the quarter.

Bonds were the worst performers in both March and in the first quarter, being beaten even by cash (+1.8%) and, as has become commonplace, by inflation-linked bonds (+2.7%). All fixed interest categories were roundly trounced by equities, with the All Share Index rising 7.1% in March to take its first quarter return to 13.3%.

It certainly was not as though any great problem beset South Africa in particular; on the contrary, the bond market took heart from a continued relatively strong rand and contained inflation. Rather, we saw bonds weaken as a spillover from jitters in global markets.

Bond yields in developed countries have generally risen to levels that are around two-year highs (though in absolute, historical terms they remain relatively low).

To some extent this represented the Federal Reserve pulling US bonds kicking and screaming upwards, as it continued a steady upward move in the Fed funds rate (now, at 4.75%, well above what consensus had expected a year ago). But the developments were actually wider than just the US, and more interesting for that, as signs of more sustained economic growth in the laggards started to feed into central bank interest rate expectations.
Europe has long been considered a "dog" in terms of growth, but a spate of positive actual data and forward-looking surveys and the European Central Bank's traditional hawkishness on inflation have combined to see official interest rates rising - with more expected to come over the course of this year. But the most interesting has probably been the confirmation of a structural change in Japan: the "end of deflation" has been proclaimed, the Bank of Japan has started to withdraw "quantitative easing" and is expected to actually raise rates sometime during the second half of the year.

Despite all this, emerging markets continued to perform well. The current line of thinking seems to be that despite the gradual withdrawal of liquidity, global growth remains robust and so emerging markets - most of which export commodities or are otherwise geared to the global growth cycle - will continue to perform well.

While accepting that this is a valid argument, we also remain somewhat cautious. South Africa will continue to do well as long as the general emerging market universe is in favour, but we note that some jitters have set in with some emerging markets (Iceland, Hungary) and some developed commodity exporters (New Zealand) finding their currencies under pressure. The rand did not escape a bout of the jitters, though at the time of writing it has retraced its losses. We remain concerned that falling global risk appetite will see currency weakening of countries that are running large current account deficits and in turn potentially place upward pressure on inflation and interest rates; South Africa would fall into this grouping.

Of course, local developments are also important in determining a final value for bond yields: while global bond market movements are likely to be a primary determinant of trend, local issues will affect South African bond spreads versus other countries. Inflation, while expected to drift up further from its trough, is not expected to provide any major scares (unless there is a sharp move downwards in the rand). However, other local developments support at best an unchanged repo rate, with risks that the next move is up - a point of view now publicly espoused by Governor Mboweni. The main concern from a monetary policy perspective at present is the booming domestic economy; while higher growth is good, the extent of what we are seeing appears to be stretching the economy's capacity. Such imbalances show up in the ever-widening current account deficit; if left unchecked to grow indefinitely, this will eventually become unsustainable with a potentially messy rand (and inflation) outcome as a result.

Thus, looking at both the global and domestic backdrops, we would expect the upward drift in bond yields to continue this year.

Mark le Roux
Portfolio Manager

2005 in the fixed income market will be remembered for a relatively stable rand, and an inflation rate (despite a material increase in the petrol price) that surprised on the downside, averaging 3.9% in 2005 versus 4.3% in 2004. The subdued inflation rate allowed the Reserve Bank to reduce the repo rate by a further 50 basis points in April (bringing total rate cuts in the cycle to total 650 basis points so far). The rate cut together with low inflation, robust growth, strong fiscal performance and foreign demand for emerging market investments, resulted in bond yields falling to their lowest levels.
Although local bond yields have reached new record lows and the All Bond Index (ALBI) showed another year of gains, diminishing returns are setting in because of the base. While the ALBI returned 10.8% last year, this was in fact its weakest performance since 1998. Moreover, local bonds far underperformed a stellar equity market return of over 47% in 2005, although they did outperform cash (7.1%). Longer-dated bonds were the best performers in the ALBI: the 12+ years maturity area returned 13.3%, while the 1-3 year area returned 7.7%.
The Coronation Bond Fund produced a return of 4.8%, pretty much in line with the ALBI return of 5.0% over the past quarter. For the year the fund returned 10.9% (net of fees), marginally above the ALBI return of 10.8%. Almost half the fund's return for the year came in the fourth quarter. Inflation-linked bonds, the star bond market performer for the preceding three quarters, finally took a little bit of a breather but still managed to notch a 4.7% return in the quarter. However on a year to date basis, they handsomely outperformed both individual nominal bonds and the ALBI, producing a 17.2% return.
The bull rally in the fourth quarter gained impetus from a stronger rand and two consecutive lower than expected inflation releases. Positive underlying sentiment towards emerging markets, stronger than expected US treasury bonds and a boost from a rising gold price all added to the underlying positive sentiment in November and December. The Emerging Market Bond Index (EMBI) spread reached a new record low in December.
One of the catalysts that caused the bond market to shrug off its lethargy occurred at the beginning of November with the announcement of the take out of VenFin by foreign investor Vodafone. The transaction would result in another material foreign direct investment flow into South Africa (following the Barclays/Absa deal) and continue to provide underlying support for the rand, which we believe would positively influence the country risk premium and the resultant stronger rand will likely keep inflation well and truly in check. Applying this view to our investment strategy, we closed out the fund's underweight duration position relative to the ALBI. The purchases resulted in the fund moving from a modified duration of 3.5 to 4.95 in line with the ALBI's modified duration.
Moving the duration of the fund to neutral has so far proved to be the correct decision as the long-dated area of the market has rallied in excess of 60 basis points since the purchases were made.
Statements from MPC meetings in October and December were worlds apart, with the Governor in October adopting a decidedly hawkish stance and warning of having to raise short rates if second round effects of a rampant oil price fed through to inflation, while in December after CPIX had surprised on the downside for the previous two months, his MPC statement took a much more dovish stance.
The yield curve experienced a fair degree of flattening during the fourth quarter as the shorter-dated area was held in check by static money market rates while demand for long-dated bonds on a benign inflation outlook continued unabated. The spread between the 10-year and three-year bonds narrowed from 53 basis points at the beginning of the quarter to 26 basis points at year end.
During the quarter corporate credit issuance continued at a rapid pace as a number of issuers saw value in locking in funding at the current low levels of rates. Credit spreads continue to trade at tight margins, and we tended to avoid the majority of new issuance as we did not feel that the spread was compensating the investor sufficiently for the risk being taken. The exception was the corporate bond issued by Old Mutual which we felt provided sufficient value and we established a holding in that bond in the fund. Credit exposure continues to run at a reasonably conservative 16% of the fund.
Looking forward, both global and domestic backdrops are likely to remain benign for bonds, but not as good as last year. We would expect an upward drift in bond yields this year. We would also like to emphasise the point made earlier about 'diminishing returns'. We think bond yields could be close to their lows at the moment, but even if they do fall further, this will be off an already low base and bonds will be hard pressed to match the returns seen in preceding years. For a South African investor, it is also unlikely that bond returns would be able to match equity returns, even if those are (as expected) significantly below the heady returns of last year.

Mark le Roux
Portfolio Manager